Low-overhead sparse beam direction determination method and device
By transmitting irregular beams and constructing a sparse model in a 5G MIMO system, and using amplitude information for iterative solution, the high computational overhead and complexity of traditional beam training methods are solved, achieving efficient and accurate beam direction determination.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-25
- Publication Date
- 2026-03-27
AI Technical Summary
In 5G MIMO systems, traditional beam training methods require comprehensive signal measurements of all possible beam directions, resulting in high computational overhead, increased time delay and resource consumption. Furthermore, they rely on precise phase information, making it difficult to reduce the number of measurements and computational complexity while maintaining training accuracy.
By transmitting irregular beams in M time slots, the measured amplitude gain at the receiver is obtained, a sparse model is constructed, and the directions of K target beams are determined by iterative solution. The amplitude information is used to avoid dependence on phase information, and sparse optimization and alternating minimization algorithms are used to reduce computational complexity.
While reducing the number of measurements and resource consumption, it improves the accuracy and robustness of beam training, reduces algorithm complexity, and enhances the ability to adapt to timing errors and phase inaccuracies.
Smart Images
Figure CN121750031A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of communication technology, and in particular to a method and apparatus for determining the direction of a low-overhead sparse beam. Background Technology
[0002] With the continuous development of 5G mobile communication technology, Multiple-Input Multiple-Output (MIMO) systems have become a key technology for improving communication performance and increasing network capacity, because MIMO can provide large-scale parallel transmission with limited spectrum resources. However, in MIMO systems, how to quickly and accurately determine the optimal beam direction is a crucial aspect of improving system performance.
[0003] In practical 5G MIMO systems, beam training is the process used to determine the optimal direction for signal transmission, typically accomplished by scanning multiple directions. This process is crucial for system throughput and reliability. However, beam training often faces a series of challenges, including computational overhead, time latency, and resource consumption, when performing large-scale direction searches. Traditional beam training methods usually require comprehensive signal measurements of all possible beam directions, especially in high-frequency millimeter-wave communication, where the number of measurements and the amount of data increase significantly, thus burdening the system. Furthermore, due to timing errors and phase inaccuracies in 5G systems, traditional beam training methods often rely on precise phase information to ensure measurement accuracy, which inherently increases algorithm complexity and computational load. Therefore, how to reduce the number of measurements and computational overhead while maintaining training accuracy has become an urgent problem to be solved in beam training technology. Summary of the Invention
[0004] This invention provides a low-overhead sparse beam direction determination method and apparatus to solve the technical problems existing in the prior art.
[0005] This invention provides a low-overhead sparse beam direction determination method, applied to a communication system, wherein the communication system includes P beam directions to be observed, and includes the following steps: The transmitter is controlled to transmit irregular beams in M time slots, and the measured amplitude gain observed by the receiver is obtained, where M is less than P; Using the weight matrix and gain matrix corresponding to K of the P observed beam directions as variables, and with the objective of minimizing the difference between the measured amplitude gain and the predicted amplitude gain, a sparse model is constructed; the predicted amplitude gain is obtained based on the weight matrix and gain matrix corresponding to the K beam directions. The sparse model is solved through K iterations to obtain K target beam directions; wherein, in each iteration, the candidate beam direction with the largest correlation coefficient in the current iteration is determined as the target beam direction in the current iteration.
[0006] According to the low-overhead sparse beam orientation determination method provided by the present invention, the target beam orientation in the current iteration round is obtained through the following steps: Determine the candidate beam directions among the P beam directions to be observed in the current iteration round, excluding all target beam directions determined before the current iteration round; For each candidate beam direction, determine the temporary gain matrix corresponding to the candidate beam direction and all the target beam directions determined before the current iteration round; Based on the temporary gain matrix and the weights of the M time slots corresponding to the candidate beam direction, the correlation coefficient corresponding to the candidate beam direction is determined. The candidate beam direction with the highest correlation coefficient is determined as the target beam direction in the current iteration round.
[0007] According to a low-overhead sparse beam orientation determination method provided by the present invention, determining the temporary gain matrix corresponding to the candidate beam orientation and all determined target beam orientations before the current iteration round includes: Determine the temporary weight matrix corresponding to the candidate beam direction and all the target beam directions determined before the current iteration round; Based on the temporary weight matrix and the measured amplitude gain, the temporary gain matrix corresponding to the candidate beam direction and all the target beam directions determined before the current iteration is determined by the alternating minimization algorithm.
[0008] According to the low-overhead sparse beam direction determination method provided by the present invention, before the control transmitter transmits irregular beams in M time slots respectively, the method further includes: Construct a complex-valued observation matrix corresponding to the P beam directions to be observed under M time slots; the element in the m-th row and p-th column of the complex-valued observation matrix represents the weight of the irregular beam to the p-th beam direction to be observed under the m-th time slot.
[0009] According to the low-overhead sparse beam direction determination method provided by the present invention, before performing K iterations of solving the sparse model, the method further includes: Initialize the support set used to record the determined target beam direction as an empty set; The initialization process uses an empty set to store the weight matrix corresponding to the determined target beam direction. Initialize the first iteration counter and the second iteration counter to 1. The maximum value of the first iteration counter is K, and the maximum value of the second iteration counter is P. Initialize the maximum correlation coefficient and the index of the target beam direction to 0.
[0010] According to the low-overhead sparse beam orientation determination method provided by the present invention, after solving the sparse model K times to obtain K target beam orientations, the method further includes: Determine the target gain corresponding to each target beam direction; The K target beam directions are sorted in descending order of target gain.
[0011] The present invention also provides a low-overhead sparse beam direction determination device, applied to a communication system, wherein the communication system includes P beam directions to be observed, including: The control module is used to control the transmitter to transmit irregular beams in M time slots and obtain the measured amplitude gain observed by the receiver, where M is less than P; The model building module is used to construct a sparse model with the weight matrix and gain matrix corresponding to K of the P observed beam directions as variables, and with the objective of minimizing the difference between the measured amplitude gain and the predicted amplitude gain; the predicted amplitude gain is obtained based on the weight matrix and gain matrix corresponding to the K beam directions. The model iteration module is used to solve the sparse model through K iterations to obtain K target beam directions; wherein, in each iteration round, the candidate beam direction with the largest correlation coefficient in the current iteration round is determined as the target beam direction in the current iteration round.
[0012] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the program to implement the low-overhead sparse beam direction determination method as described above.
[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the low-overhead sparse beam direction determination method as described above.
[0014] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the low-overhead sparse beam direction determination method as described above.
[0015] This invention provides a low-overhead sparse beam direction determination method and apparatus, applied to a communication system comprising P beam directions to be observed. The method includes controlling the transmitter to transmit irregular beams in M time slots and acquiring the measured amplitude gain observed by the receiver, where M is less than P; constructing a sparse model using the weight matrices and gain matrices corresponding to K beam directions among the P beam directions as variables, with the objective of minimizing the difference between the measured amplitude gain and the predicted amplitude gain; the predicted amplitude gain is obtained based on the weight matrices and gain matrices corresponding to the K beam directions; and solving the sparse model through K iterations to obtain K target beam directions; wherein, in each iteration, the candidate beam direction with the highest correlation coefficient in the current iteration is determined as the target beam direction in the current iteration. This invention reduces the measurement overhead and resource consumption required for beam training by performing measurements in M time slots, which is less than the total number of beams to be observed, P. Secondly, by using only the measured amplitude gain observed at the receiver to construct the model and solve for the problem, the reliance on precise phase information is avoided, thus reducing the computational complexity of the algorithm and enhancing the robustness of the method to timing errors and phase inaccuracies in the system. Finally, by constructing an optimization model with the objective of minimizing the difference between the predicted and measured amplitude gain, and employing an iterative solution method, only the correlation coefficient of the current candidate beam needs to be calculated each time. This avoids the comprehensive measurement and complex matrix operations for all beam directions in traditional methods, thereby achieving high-precision determination of the target beam direction while significantly reducing the number of measurements and eliminating reliance on phase information. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0017] Figure 1 This is one of the flowcharts illustrating the low-overhead sparse beam direction determination method provided by the present invention.
[0018] Figure 2 This is the second flowchart of the low-overhead sparse beam direction determination method provided by the present invention.
[0019] Figure 3 This is a schematic diagram of the low-overhead sparse beam direction determination device provided by the present invention.
[0020] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation
[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.
[0022] It should be noted that in the description of this invention, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. The terms "upper," "lower," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the system or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.
[0023] The terms "first," "second," etc., used in this invention are used to distinguish similar objects, not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, a first object can be one or more. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.
[0024] Currently, traditional methods for 5G MIMO beam training typically rely on comprehensive search strategies, requiring the transmitter or receiver to perform accurate signal measurements in all possible directions. This inevitably increases measurement time, computational complexity, and system resource consumption in large-scale antenna array systems. Furthermore, due to timing errors and signal phase uncertainties, traditional methods often require complete phase information for accurate beam pointing, leading to high computational overhead in practical applications.
[0025] Based on this, this invention proposes a low-overhead sparse beam direction determination method for a communication system. This system includes a transmitter and a receiver. The transmitter is equipped with an antenna, for example, a uniform planar array (UPA), and operates through a single radio frequency chain (RF chain). The system has P pre-defined beam directions to be observed, while the target beam directions for strong signal paths are sparse; in this embodiment, we assume there are K such directions, where K is much smaller than P.
[0026] This invention reduces the computational overhead of 5G MIMO beam training by combining Gaussian matrices and sparsity theory, using only amplitude information. Specifically, by performing M scans with a single radio frequency link equipped with an N×N UPA, the K beam directions with the highest gain can be effectively recovered.
[0027] This invention provides a low-overhead sparse beam direction determination method. Figure 1 This is a flowchart illustrating the low-overhead sparse beam direction determination method provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes the following steps 110, 120 and 130.
[0028] Step 110: Control the transmitter to transmit irregular beams in M time slots respectively, and obtain the measured amplitude gain observed by the receiver, where M is less than P.
[0029] Here, irregular beams refer to non-directional probe beams that can simultaneously cover multiple directions to be observed. Compared to traditional methods that transmit directional beams pointing in a single direction one by one, this embodiment transmits M irregular beams, which can collect information covering all directions of space within M time slots, thereby greatly reducing the number of time slots required for measurement.
[0030] The measured amplitude gain refers to the signal amplitude gain observed at the receiver in M time slots, denoted as a vector. , .
[0031] Furthermore, before controlling the transmitter to transmit irregular beams, the method further includes: constructing complex-valued observation matrices corresponding to the directions of P beams to be observed under M time slots; the element in the m-th row and p-th column of the complex-valued observation matrix represents the weight of the irregular beam to the p-th direction to be observed under the m-th time slot.
[0032] Here, the complex-valued observation matrix can be accurately calculated and constructed based on the physical structure of the antenna array and the characteristics of the emitted irregular beam. .
[0033] Specifically ,in The complex-valued observation matrix constructed above represents the weights of the UPA in different directions each time it emits an irregular beam. ( () is the number of observation directions, which is usually less than or equal to the size of UPA. It is a sparse complex matrix. Indicates the first Gain in each direction, It is determined by the relative positions of the receiver and the transmitter. It is Additive White Gaussian Noise (AWGN). , This represents the noise standard deviation.
[0034] It should be noted that since timing errors can cause phase changes, this embodiment only uses the amplitude information of the signal and ignores the phase information in order to avoid interference from unstable factors and enhance robustness.
[0035] Step 120: Using the weight matrix and gain matrix corresponding to K of the P observed beam directions as variables, and with the objective of minimizing the difference between the measured amplitude gain and the predicted amplitude gain, a sparse model is constructed; the predicted amplitude gain is obtained based on the weight matrix and gain matrix corresponding to the K beam directions.
[0036] The sparse model is based on compressed sensing theory. Specifically, since the target beam directions are sparse, the positions and gains of the K target beams can be recovered from a small number of measurements by solving a sparse optimization problem.
[0037] The objective function of the sparse model can be specifically expressed as: .in, This is the measured amplitude gain. It is the weight matrix to be solved, corresponding to the directions of the K target beams. It is the gain matrix to be solved for the K target beam directions. This refers to the predicted amplitude gain calculated based on the variables. By minimizing the difference between the measured amplitude gain and the predicted amplitude gain, K beam directions and their gains are found.
[0038] in, , The solved number is Gain in each beam direction It can be seen as from Extract the one with the highest gain It consists of the gain in each beam direction. It can be seen as from Extract the corresponding sample with the largest gain. It consists of column vectors of beam directions.
[0039] Step 130: Solve the sparse model for K iterations to obtain K target beam directions; wherein, in each iteration, the candidate beam direction with the largest correlation coefficient in the current iteration is determined as the target beam direction in the current iteration.
[0040] It should be understood that since directly solving this model is a complex non-convex problem, this embodiment of the invention employs a two-layer iterative greedy algorithm. Specifically, K iterations are performed, with each iteration identifying a target beam direction. In each iteration, the beam direction with the highest correlation coefficient is determined from all unselected candidate beam directions and identified as the target beam direction found in the current round. Here, the correlation coefficient is a key indicator used to characterize the degree of matching of a candidate direction. The larger the correlation coefficient, the more likely the candidate direction is to be one of the target beam directions.
[0041] Furthermore, in this embodiment, to ensure the correct execution of the iterative solution process, the following initialization operation needs to be performed before starting the Kth iteration: Initialize the support set used to record the determined target beam direction. This is an empty set, and the support set is used to record the index of the target beam direction that has been determined in each iteration; Initialize the weight matrix This is an empty set; the matrix is used to store the weight vector corresponding to the determined target beam direction, i.e., the complex-valued observation matrix. The columns in; Initialize the iteration counters: Initialize the first iteration counter k=1, with a maximum value of K, which is used to control the outer loop (finding a total of K directions). Initialize the second iteration counter n=1, with a maximum value of P, which is used for the inner loop (traversing all candidate beam directions). Initialize intermediate variables: Initialize the variable used to record the maximum correlation coefficient to 0, and set the sign of the variable used to record the target beam direction index in the current iteration round to 0.
[0042] In some embodiments, the target beam direction in the current iteration round is obtained through the following steps: Determine the candidate beam directions among the P beam directions to be observed in the current iteration round, excluding all target beam directions determined before the current iteration round; For each candidate beam direction, determine the temporary gain matrix corresponding to the candidate beam direction and all the target beam directions determined before the current iteration round; Based on the temporary gain matrix and the weights of the M time slots corresponding to the candidate beam direction, the correlation coefficient corresponding to the candidate beam direction is determined. The candidate beam direction with the highest correlation coefficient is determined as the target beam direction in the current iteration round.
[0043] First, determine the set of candidate beam directions for the current iteration t. This set includes all P beam directions to be observed, excluding all target beam directions determined in the previous t-1 iterations. In this embodiment, the set of determined target beam directions can be recorded in the support set. Therefore, the candidate beam direction in the current iteration round t is... .
[0044] For each candidate beam direction n, it is assumed that it, together with all previously determined target beam directions, constitutes the true beam set. Based on this assumption, a temporary gain matrix is calculated. Here, the temporary gain matrix reflects the amplitude gain estimates for each beam direction under this assumption. Based on the temporary gain matrix... And the weights of the candidate beam direction n in the M time slots, i.e., the column vectors corresponding to the candidate beam direction n in the observation matrix A. Calculate the correlation coefficient corresponding to the candidate beam direction n. The correlation coefficient can be specifically expressed as: .
[0045] Iterate through all candidate beam directions and calculate their respective correlation coefficients. Select the candidate beam direction with the highest correlation coefficient as the target beam direction determined in the current t-th iteration. Then, add the index of this target beam direction to the support set and update it to I(t) for use in the next iteration.
[0046] In this embodiment, the above iterative method ensures that each step makes a locally optimal choice, thereby gradually approaching the global optimal solution and finally accurately finding the K target beam directions.
[0047] In some embodiments, determining the temporary gain matrix corresponding to the candidate beam direction and all previously determined target beam directions includes: Determine the temporary weight matrix corresponding to the candidate beam direction and all the target beam directions determined before the current iteration round; Based on the temporary weight matrix and the measured amplitude gain, the temporary gain matrix corresponding to the candidate beam direction and all the target beam directions determined before the current iteration is determined by the alternating minimization algorithm.
[0048] First, construct the temporary weight matrix for the current iteration round t. The temporary weight matrix consists of two parts: one part is the weight matrix corresponding to all determined target beam directions up to the current iteration t. The other part is the weight corresponding to the candidate beam direction n. .
[0049] Construct the temporary weight matrix Then, the optimization problem is solved using an alternating minimization algorithm. Obtain the temporary gain matrix Here, and These are all temporary variables at iteration t. In this embodiment, we use alternating methods to fix some variables while optimizing others to solve for the temporary gain matrix. .
[0050] loop iteration After that, according to and The maximum gain can be obtained One direction and The estimated value , The obtained K target beam directions and their corresponding target gains are paired, and then sorted in descending order according to the magnitude of the target gain. The communication system can not only know which K signal paths are effective, but also their signal strength ranking. This allows the system to prioritize the beam with the strongest signal for data transmission, thereby optimizing communication link quality and user experience.
[0051] This embodiment's low-overhead sparse beam orientation determination method reduces the measurement overhead and resource consumption required for beam training by performing measurements over M time slots, which is less than the total number of beams P to be observed. Secondly, it utilizes only the measured amplitude gain observed at the receiver to construct the model and solve for the beam, avoiding reliance on precise phase information, thus reducing the algorithm's computational complexity and enhancing its robustness to timing errors and phase inaccuracies in the system. Finally, by constructing an optimization model that minimizes the difference between the predicted and measured amplitude gain and employing an iterative solution, only the correlation coefficient of the current candidate beam needs to be calculated each time. This avoids the comprehensive measurement and complex matrix operations required for all beam orientations in traditional methods, thereby achieving high-precision target beam orientation determination with a significantly reduced number of measurements and without relying on phase information.
[0052] For ease of understanding, Figure 2 This is a second schematic flowchart of the low-overhead sparse beam direction determination method provided in this embodiment of the invention, as shown below. Figure 2 As shown, the low-overhead sparse beam direction determination method provided in this embodiment of the invention includes the following steps: First, input the known parameters: the measured amplitude gain of the receiver. Complex-valued observation matrix Number of target beam directions K; Total number of beam directions to be observed P; Initialize variables: support set Initialize the weight matrix for an empty set (no beam direction selected). Empty set (unsampled complex-valued observation matrix) (any column vector), the first iteration counter k=1 (currently, the direction of the kth target beam needs to be determined); Next, the outer iteration proceeds for a total of K rounds, with each round determining one target beam direction. The specific iteration process is as follows: The iteration ends when k > K; when k ≤ K, the following process is executed: Initialize inner variables: maximum correlation coefficient M cor =0, current target beam direction index equals 0, second iteration counter n=1; In each outer iteration, if If n is not in the selected support set and n≤P, then execute the following inner loop: Constructing a temporary weight matrix = [ Solve using the Altmin algorithm. Calculate the correlation coefficient of the current candidate beam direction n; If the correlation coefficient of the current candidate beam direction n > M cor Update M cor Let n = n+1, and repeat the inner loop until all unselected candidate beam directions are traversed, and set the final correlation coefficient to M. cor The candidate beam direction n is added to the support set I to obtain From the complex-valued observation matrix Extract the column vector in that direction and update .
[0053] After the outer iteration is completed, based on the final support set and The temporary gain matrix is filled with the gain of the corresponding K target beam directions to obtain the estimated value x̂ of x, and the K maximum gain directions and their corresponding gains are output.
[0054] The low-overhead sparse beam direction determination device provided by the present invention is described below. The low-overhead sparse beam direction determination device described below can be referred to in correspondence with the low-overhead sparse beam direction determination method described above.
[0055] The low-overhead sparse beam direction determination device of this invention, such as Figure 3 As shown, it includes the following modules: The control module 310 is used to control the transmitter to transmit irregular beams in M time slots respectively, and to obtain the measured amplitude gain observed by the receiver, where M is less than P; The model building module 320 is used to construct a sparse model with the weight matrix and gain matrix corresponding to K of the P beam directions to be observed as variables, and with the objective of minimizing the difference between the measured amplitude gain and the predicted amplitude gain; the predicted amplitude gain is obtained based on the weight matrix and gain matrix corresponding to the K beam directions. The model iteration module 330 is used to solve the sparse model through K iterations to obtain K target beam directions; wherein, in each iteration round, the candidate beam direction with the largest correlation coefficient in the current iteration round is determined as the target beam direction in the current iteration round.
[0056] The low-overhead sparse beam orientation determination device in this embodiment reduces the measurement overhead and resource consumption required for beam training by performing measurements in M time slots, which is less than the total number of beams P to be observed. Secondly, it constructs and solves the model using only the measured amplitude gain observed at the receiver, avoiding reliance on precise phase information, thus reducing the computational complexity of the algorithm and enhancing the method's robustness to timing errors and phase inaccuracies in the system. Finally, by constructing an optimization model that minimizes the difference between the predicted and measured amplitude gain and employing an iterative solution, only the correlation coefficient of the current candidate beam needs to be calculated each time. This avoids the comprehensive measurement and complex matrix operations required for all beam orientations in traditional methods, thereby achieving high-precision determination of the target beam orientation with a significantly reduced number of measurements and without relying on phase information.
[0057] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4As shown, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communication bus 440. The processor 410, communications interface 420, and memory 430 communicate with each other via the communication bus 440. The processor 410 can call logical instructions from the memory 430 to execute a low-overhead sparse beam direction determination method, applied to a communication system including P beam directions to be observed. The method includes: The transmitter is controlled to transmit irregular beams in M time slots, and the measured amplitude gain observed by the receiver is obtained, where M is less than P; Using the weight matrix and gain matrix corresponding to K of the P observed beam directions as variables, and with the objective of minimizing the difference between the measured amplitude gain and the predicted amplitude gain, a sparse model is constructed; the predicted amplitude gain is obtained based on the weight matrix and gain matrix corresponding to the K beam directions. The sparse model is solved through K iterations to obtain K target beam directions; wherein, in each iteration, the candidate beam direction with the largest correlation coefficient in the current iteration is determined as the target beam direction in the current iteration.
[0058] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc., any medium capable of storing program code.
[0059] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program that can be stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is able to execute the low-overhead sparse beam direction determination method provided by each of the above methods, applied to a communication system, the communication system including P beam directions to be observed, the method comprising: The transmitter is controlled to transmit irregular beams in M time slots, and the measured amplitude gain observed by the receiver is obtained, where M is less than P; Using the weight matrix and gain matrix corresponding to K of the P observed beam directions as variables, and with the objective of minimizing the difference between the measured amplitude gain and the predicted amplitude gain, a sparse model is constructed; the predicted amplitude gain is obtained based on the weight matrix and gain matrix corresponding to the K beam directions. The sparse model is solved through K iterations to obtain K target beam directions; wherein, in each iteration, the candidate beam direction with the largest correlation coefficient in the current iteration is determined as the target beam direction in the current iteration.
[0060] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a low-overhead sparse beam direction determination method provided by each of the above methods, applied to a communication system including P beam directions to be observed, the method comprising: The transmitter is controlled to transmit irregular beams in M time slots, and the measured amplitude gain observed by the receiver is obtained, where M is less than P; Using the weight matrix and gain matrix corresponding to K of the P observed beam directions as variables, and with the objective of minimizing the difference between the measured amplitude gain and the predicted amplitude gain, a sparse model is constructed; the predicted amplitude gain is obtained based on the weight matrix and gain matrix corresponding to the K beam directions. The sparse model is solved through K iterations to obtain K target beam directions; wherein, in each iteration, the candidate beam direction with the largest correlation coefficient in the current iteration is determined as the target beam direction in the current iteration.
[0061] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0062] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in each of the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of each embodiment of the present invention.
Claims
1. A low-overhead sparse beam direction determination method, applied to a communication system, wherein the communication system includes P beam directions to be observed, characterized in that, The method comprises the steps of: controlling the transmitting end to transmit irregular beams in M time slots respectively, and obtaining measured amplitude gains observed by the receiving end, M being less than P; taking the weight matrix and the gain matrix corresponding to K beam directions in P to-be-observed beam directions as variables, and taking the minimum difference between the measured amplitude gains and predicted amplitude gains as a target, a sparse model is constructed; the predicted amplitude gains are obtained according to the weight matrix and the gain matrix corresponding to the K beam directions; K times of iteration solving are performed on the sparse model to obtain K target beam directions; wherein in each iteration round, a candidate beam direction with the largest correlation coefficient in the current iteration round is determined as the target beam direction in the current iteration round.
2. The low-oversight sparse beam direction determination method of claim 1, wherein, the target beam direction in the current iteration round is obtained through the following steps: determining candidate beam directions in the P to-be-observed beam directions in the current iteration round except for all the target beam directions determined before the current iteration round; for each candidate beam direction, determining a temporary gain matrix corresponding to the candidate beam direction and all the target beam directions determined before the current iteration round; determining the correlation coefficient corresponding to the candidate beam direction according to the temporary gain matrix and the weight in the M time slots corresponding to the candidate beam direction; determining the candidate beam direction with the largest correlation coefficient as the target beam direction in the current iteration round.
3. The low-oversight sparse beam direction determination method of claim 2, wherein, the determination of the temporary gain matrix corresponding to the candidate beam direction and all the target beam directions determined before the current iteration round comprises: determining a temporary weight matrix corresponding to the candidate beam direction and all the target beam directions determined before the current iteration round; determining the temporary gain matrix corresponding to the candidate beam direction and all the target beam directions determined before the current iteration round through an alternating minimization algorithm according to the temporary weight matrix and the measured amplitude gains.
4. The low-oversight sparse beam direction determination method of claim 1, wherein, Before the control of the transmitting end to transmit irregular beams in M time slots respectively, the method further comprises the steps of: constructing a complex observation matrix corresponding to the P to-be-observed beam directions in the M time slots; an element in the mth row and the pth column in the complex observation matrix represents the weight of the irregular beam in the mth time slot to the pth to-be-observed direction.
5. The low-oversight sparse beam direction determination method of claim 1, wherein, Before the K times of iteration solving of the sparse model, the method further comprises the steps of: initializing a support set for recording the determined target beam directions as an empty set; initializing a weight matrix for storing the determined target beam directions as an empty set; initializing a first iteration counter and a second iteration counter to be equal to 1, the maximum value of the first iteration counter being K, and the maximum value of the second iteration counter being P; initializing the maximum correlation coefficient and the index of the target beam direction to be 0.
6. The low-oversight sparse beam direction determination method according to any one of claims 1 to 5, characterized in that, After the K times of iteration solving of the sparse model to obtain the K target beam directions, the method further comprises the steps of: determining a target gain corresponding to each target beam direction; sorting the K target beam directions in the order from large to small according to the target gains.
7. A low-overhead sparse beam direction determination device, applied to a communication system, wherein the communication system includes P beam directions to be observed, characterized in that, The method comprises the steps of: a control module, configured to control the transmitting end to transmit irregular beams in M time slots respectively, and obtain measured amplitude gains observed by the receiving end, M being less than P; A model construction module is configured to construct a sparse model by taking weight matrices and gain matrices corresponding to K beam directions in P to-be-observed beam directions as variables and taking a minimum difference between the measured amplitude gain and the predicted amplitude gain as a target; The predicted amplitude gain is obtained according to the weight matrices and the gain matrices corresponding to the K beam directions; A model iteration module is configured to perform K times of iteration on the sparse model to obtain the K target beam directions; wherein, in each iteration round, a candidate beam direction with a maximum correlation coefficient in the current iteration round is determined as a target beam direction in the current iteration round.
8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and running on the processor, characterized in that, The processor executes the computer program to implement the low-overhead sparse beam direction determination method in any one of claims 1 to 6. 9.A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the low-overhead sparse beam direction determination method in any one of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the low-overhead sparse beam direction determination method in any one of claims 1 to 6.